Infection by ecotropic murine leukemia viruses (MLV) shows several characteristic anomalies, including entry by an atypical endocytic pathway, as well as entry by cell surface fusion and a much slower rate of membrane fusion. Lack of information on the molecular basis for these anomalies has hampered an understanding of envelope protein-mediated viral pathogenesis and improvements to the design and development of retroviral vectors for gene therapy. Based on new preliminary data, we propose a model for the post-binding events. The model consists of two novel hypotheses: First, that the number of receptors and their membrane micro-domain localization influences the kinetics of virus-cell membrane fusion. We show preliminary data that the time to onset of fusion and the rate of fusion differs markedly between NIH 3T3 and XC cells. Receptors are fewer and do not cluster on XC cells as on other cell types. In addition, XC cells are more sensitive to cholesterol depletion than NIH 3T3 cells, together suggesting that the differences in fusion kinetics may result from lower receptor levels and differences in their membrane localization. We will test that concept in Aim 1. The second hypothesis is that the major constraint holding Env in the metastable conformation is the continuity of the peptide chain of SU and that cleavage at specific sites by host cell proteases releases the constraint so that receptor-induced changes give the stable fusion-competent conformation. In new preliminary studies, we show evidence that cathepsins are good candidates for triggering proteases and propose candidate cleavage sites. Aim 2 will complete identification of candidate proteases and the sites cleaved, then determine if cleavage at these sites is relevant to infection. Aim 3 will focus on the molecular mechanisms of envelope protein (Env) function in membrane fusion. The step in membrane fusion that is blocked by Env fusion mutants will be identified using the virus- and cell-cell fusion assays and a detailed functional map of the Env will be built by assigning mutants to genetic complementation groups using a subunit trans-complementation assay.